Life of Research
Upload
Life of Research
Sign inCreate account
Appearance

Getting found and cited

How to get your research indexed in Google Scholar

Updated 5 min read By Life of Research

Short answer

Google Scholar indexes scholarly documents it can crawl and parse: one paper per page, the title, authors and publication date in citation meta tags, the abstract or full text visible without signing in, and a PDF with searchable text no larger than 5 MB, reachable through ordinary links. Put the paper on a site that meets those rules, such as a repository, link to it, and allow time. Scholar gives no guarantee of inclusion.

How does Google Scholar find papers?

Google Scholar finds papers by crawling the web, like Google's main search, but it keeps only documents that look like scholarly work: journal and conference papers, technical reports, theses and dissertations, preprints, postprints and abstracts. It reads bibliographic details from metadata tags in each page and from the document itself, then groups the publisher's version, preprints and repository copies of the same paper into one cluster, shown as "All versions".

Scholar publishes inclusion guidelines for website owners. Individual authors are told something simpler: put the paper on a website and link to it from your publications page, and the crawler should normally find it. Either way, the technical rules below decide whether a page can be indexed.

What does a page need for Google Scholar to index it?

According to Scholar's inclusion guidelines, an indexable paper needs:

  • Its own page or file. Each paper and each abstract must be in a separate HTML page or PDF; Scholar cannot index several papers in one PDF or several abstracts on one page.
  • A visible abstract or full text. Readers arriving from search must see the complete abstract without signing in, accepting disclaimers, dismissing pop-ups, clicking or scrolling.
  • Citation metadata. At minimum the title, the full name of at least the first author and the year of publication. Scholar supports Highwire Press tags (such as citation_title), bepress tags and PRISM tags; Dublin Core is accepted only as a last resort because it cannot express journal, volume, issue and pages unambiguously.
  • A link to the full text in a citation_pdf_url tag.
  • A suitable PDF. HTML or PDF only, with searchable text, no larger than 5 MB.
  • A crawl path. Every paper reachable from the home page through at most ten ordinary HTML links, with a listing by date for large sites, and robots.txt must not block the papers or the listing pages.
Highwire Press tags in a page's head (invented example)
<meta name="citation_title" content="Seasonal variation in river temperature in the Upper Tana basin">
<meta name="citation_author" content="Okafor, Chidinma N.">
<meta name="citation_author" content="Lindqvist, Maja">
<meta name="citation_publication_date" content="2025/03/14">
<meta name="citation_pdf_url" content="https://example.org/papers/river-temperature.pdf">

Does the layout of the PDF matter?

Yes. Scholar parses the PDF itself, so the first page should look like a paper. Its guidelines ask for the title to be the largest text at the top of the first page (at least 24 pt in a PDF), the authors listed right before or after the title in a slightly smaller font (16 to 23 pt), and the references in a section whose heading, such as "References" or "Bibliography", stands on a line by itself. They also advise against Type 3 fonts, which often lose the size and encoding information the parser needs.

The metadata tags and the PDF should agree. If the title in the tags differs from the title on the first page, or the PDF is a scanned image with no text layer, the paper may be indexed incorrectly or not at all. Run optical character recognition on scanned documents before depositing, and compress or split PDFs larger than 5 MB.

Why might a paper not appear in Google Scholar?

The usual causes are technical, and most can be fixed:

  • the PDF is scanned, has no text layer, or is larger than 5 MB;
  • the page needs a login, a cookie wall or JavaScript interaction before the abstract appears;
  • there are no citation meta tags, or they disagree with the PDF;
  • the paper sits several clicks deep behind search forms, with no plain listing that links to it;
  • the site mixes papers with large amounts of non-scholarly content;
  • several different documents share an identical title, so Scholar treats them as duplicates;
  • not enough time has passed.

How long does Google Scholar indexing take?

For a site Scholar already indexes, its guidelines say new papers are normally added several times a week, while changes to papers already included usually take six to nine months to show. For a paper on an individual author's site, Scholar says its robots should normally find it within several weeks.

Inclusion is not guaranteed for any paper or any site. Scholar decides which sites to index and does not offer a submission form for individual papers.

Should you create a Google Scholar profile?

Yes, if you want a public list of your papers with citation counts. A Google Scholar profile collects the articles Scholar has attributed to you, can update itself as new ones are found, and can be made public.

You can add a missing article to your profile by hand, but that only describes the paper; it does not give Scholar a crawlable copy of the full text. The reliable fix for a missing paper is still a public page and PDF that meet the requirements above.

How does Life of Research help?

Record pages on Life of Research are built to meet these technical requirements:

  • each record has its own page with the full abstract visible to everyone, no login and no interstitials;
  • pages carry Highwire Press tags for title, every author, ORCID iDs, publication date, DOI and, for journal articles, the journal, volume, issue and pages;
  • the first PDF attached is declared as the full text in citation_pdf_url, served from the same site;
  • every record is reachable through the archive by year and month in a few links, and papers are kept under their own addresses (/articles/, /preprints/, /theses/, /reports/) separate from datasets and software;
  • the deposit form requires an abstract of at least 150 characters, and a warning appears straight after publishing when the main PDF is larger than about 5 MB or has no text layer.

What the platform cannot do is guarantee inclusion: Scholar evaluates sites over time and makes its own decisions. Depositing in more than one place, such as your institution's repository as well, gives Scholar more than one route to your paper.

A checklist before you deposit

  • The PDF has selectable text and is no larger than 5 MB.
  • The title on the first page matches the title you enter, word for word.
  • All authors are listed in order with full names.
  • The abstract is complete and meaningful on its own.
  • The publication date is correct, and the DOI and journal details are filled in for published articles.
  • The manuscript PDF is the first file attached.
  • You link to the record from your own website, CV and profiles, which gives crawlers another path to it.

Key terms